1. Field of the Invention
This invention relates to capillary viscometers. More specifically, it relates to differential pressure capillary viscometers which may be used alone to measure the viscosity of fluids or together with a chromatograph to obtain accurate viscosity information for determining molecular weight distributions.
2. Discussion of the Prior Art
Accurate measurements of fluid viscosity are important in many industries. A capillary viscometer is often used to measure the absolute viscosity of a given fluid. For many purposes, however, it is necessary to know the relative viscosity of two fluids. Relative viscosity is often determined experimentally by measuring the absolute viscosity of each fluid separately with a capillary viscometer and then calculating the ratio.
Relative viscosity is of particular importance in polymer research and manufacturing since it can be used to measure molecular weights and to determine molecular weight distributions, which provide important information relating to the physical properties of polymers. A comparison of the viscosity behavior of two polymers of the same molecular weight, for example, is used as a measure of the degree of branching. One of the oldest means used to obtain such information is to measure the viscosity of a known concentration of a polymer in a solvent. By utilizing the ratio of the viscosity of the polymer-solvent solution, .eta..sub.p, to that of the pure solvent, .eta..sub.s, the intrinsic viscosity [.eta.] of the polymer can be calculated in accordance with following mathematical relationship:
Relative viscosity .eta..sub.r =.eta..sub.p /.eta..sub.s PA1 Specific viscosity .eta..sub.sp =.eta..sub.r -1 PA1 Inherent viscosity .eta..sub.inh =(ln.eta..sub.r)/C PA1 (where C is the polymer weight concentration and ln is the symbol for natural logarithm) PA1 passing the solvent through a first capillary tube, PA1 passing the solution through a second capillary tube connected in series with the first capillary tube, PA1 measuring separately the pressure drop across the first and second capillary tubes when each is full of flowing solvent and solution, respectively, PA1 generating signals corresponding to the pressure drop across each capillary tube, and PA1 employing said signals to measure either the intrinsic or inherent viscosity of the solute by use of logarithmic amplification wherein the viscosity measured is independent of flow rate and temperature fluctuations. PA1 a first capillary tube through which the solvent flows, PA1 a second capillary tube arranged in series with the first capillary tube and through which the solution flows, PA1 solvent supply means for supplying solvent to flow through both capillary tubes, PA1 solution supply means for supplying a sample of the solute to the solvent stream so that solution flows through the second capillary tube, PA1 means for measuring the pressure drop across each capillary tube and generating a signal responsive to each pressure drop, and PA1 logarithmic amplification means for receiving and processing the pressure drop signals for use in measuring either the intrinsic or inherent viscosity of the solute independent of the flow rate and temperature fluctuations of the solution and the solvent. PA1 K is the instrument constant which is proportional to the capillary's length l and internal diameter d as follows: l/d.sup.4 PA1 Q is the volume flow rate PA1 .eta. is the effluent liquid viscosity.
and finally ##EQU1## (where ##EQU2## is the mathematical symbol meaning limit of the quantity when the concentration C approaches zero)
Inherent and intrinsic viscosities are important polymer characterization parameters. The intrinsic viscosity, for example, provides an indication of the size of the polymer molecules. The value of [.eta.] is not a function of polymer concentration or the viscosity of the solvent medium. The value of [.eta.] for a linear polymer in a specific solvent is related to the polymer molecular weight M through the Mark-Houwink Equation: EQU [.eta.]=KM.sup..alpha.
where K and .alpha. are Mark-Houwink viscosity constants, some of which are available in polymer handbooks.
Prior art viscometers have been designed to measure viscosities in a number of ways. An early device used a single capillary of known diameter and length. Both the volume rate of flow of the solution and the pressure drop for flow through the capillary are measured. The pressure drop is usually measured by an electrical signal generated by a pressure transducer. The various viscosities are then calculated from the known geometrical parameters of the capillary.
These types of prior art viscometers have proved to be inaccurate because of signal-to-noise problems in the signal generated by the pressure transducer. Part of this problem is due to high frequency pumping noise and back pressure fluctuations. The former is caused by the reciprocating action of the high frequency pumps commonly used to move solvent and polymer solution through the viscometer. Back pressure fluctuations cause flow rate fluctuations and occur when the sample solution goes through various high resistant elements such as end-frits, valves, connectors and the capillary tube itself.
A significant part of this problem is due to actual flow rate fluctuations. These can occur for a number of reasons such as whenever the polymer sample is injected into the solvent stream. Sample injection upsets the flow rate, and noise is generated which hinders accurate viscosity measurement. Accordingly, measuring viscosity independent of flow rate fluctuations is critical to accurate viscosity readings.
Viscosity measurements are also very sensitive to temperature fluctuations. These can occur when the temperature of the solvent supply is not controlled carefully and is affected by environmental temperature changes. Sample injection can also cause temperature upsets. Measuring viscosity independent of temperature fluctuations is therefore important in obtaining accurate viscosity measurements.
An improved viscometer described in U.S. Pat. No. 3,808,877, granted May 7, 1974, to David E. Blair and assigned to the same assignee as the present application, sought to solve some of these problems. The Blair viscometer used a flow restrictor between the solvent supply and capillary to try to maintain the flow rates constant. Also, it measured relative viscosity by taking separate pressure drop measurements, first when polymer solution and then when pure solvent flowed through the capillary. In its preferred embodiment, two capillary tubes were arranged in parallel, one filled with polymer solution and the other only with solvent. In another version, the two tubes were connected in series, with pure solvent flowing through one and polymer solution through the other, and the pressure drop across each was measured. In each case, the capillary tubes had to be exactly matched so as to be identical in diameter and length. Otherwise, the pressure drops across both would not be equal for a given flow rate. If, however, the capillaries were not maintained at the identical temperature, they became "unmatched" in pressure drop, which resulted in lower sensitivity. Fluctuations in flow rate as well as temperature also adversely affected the accuracy of Blair's relative viscosity measurements.
A capillary viscometer which is similar to Blair's preferred parallel arrangement for measuring the differential pressure across a capillary bridge is described in U.S. Pat. No. 4,463,598 granted Aug. 7, 1984, to Max A. Haney. This device, as in the case of Blair's, requires the capillaries to be matched. Also, Haney, like Blair, does not compensate for flow rate and temperature fluctuations in real time when it measures the differential pressure drops. Consequently, neither Blair nor Haney provides viscosity measurements independent of flow rate and temperature fluctuations.
A more accurate viscometer is also needed in size exclusion chromatography (SEC) such as gel permeation chromatography (GPC) analysis. This technique has become widely used because of its ability to separate polymeric materials in a dilute solution according to molecular size. It utilizes columns containing porous packings which are capable of separating the molecules in a multicomponent polymer sample according to their size. The polymer components migrate through the column at different velocities and elute separately from the column at different times. The largest polymer molecules elute first and the smallest molecules last. By detecting the amount of polymer fractions in the eluant, a GPC elution curve is generated which reflects the molecular weight distribution of the multicomponent polymer sample.
In a GPC device, a detector is commonly used for providing the weight concentration profile of the elution curve. The polymer sample concentration profile is commonly obtained by using a differential refractometer. Molecular weight information is provided indirectly by the time of elution (also referred to as retention time) of the different polymer components in the sample. By using standardized correlations, which are often not available, the molecular weight distribution of the polymer sample could be calculated.
One major drawback with this type of instrument is that it does not directly measure the molecular weight of polymer molecules as they elute from the GPC colums. Also, there is no calibration of the GPC peak retention times with the polymer molecular weight. Rather, it is necessary to assume some relationship between retention times and molecular weights over a wide range of different polymer structures. Quite obviously, a detector which will provide a means for measuring molecular weight directly is preferable.
A continuous capillary type viscometer was proposed for GPC analysis by A. C. Ouano in J. Polym. Sci. Part A-1, 10 2169 (1972). A single capillary tube was placed in series with a concentration detector such as a differential refractometer at the exit end of the GPC column. As liquid continuously flowed through the capillary, the pressure drop across the capillary was measured and recorded. When a polymer solution of higher viscosity than the solvent reached the capillary, a peak in the .DELTA.P recording trace was detected. Variations of this type of GPC-viscosity detector are described in U.S. Pat. No. 3,837,217, granted Sept. 24, 1974, to W. W. Schulz and U.S. Pat. No. 4,286,457, granted Sept. 1, 1981, to H. W. Johnson.
These GPC-viscosity detectors, however, remain inaccurate because the pressure drop .DELTA.P signals are still subject to flow rate and temperature fluctuations. While the improvements described in Blair in U.S. Pat. No. 3,808,877 might help somewhat in GPC analysis as suggested by Haney in U.S. Pat. No. 4,463,598, there remains a need for an accurate means for measuring viscosity which is truly independent of fluctuations in flow rate and temperature. The present invention overcomes the problems associated with the prior art devices by eliminating the need for matched capillaries and by eliminating the dependence of the relative viscosity measurement on fluctuations in flow rate and temperature.